What is it about?
Artificial Intelligence (AI) is becoming more common in how we manage electricity and power grids. This paper looks at how AI can help make energy systems more efficient, reliable, and sustainable—but it also explores the challenges that come with using AI. These include concerns about transparency, fairness, data privacy, and whether AI can work well with older power grid technology. By reviewing over 60 academic studies, we identify the key issues and suggest ways policymakers and engineers can use AI more responsibly in the energy sector.
Featured Image
Photo by Markus Winkler on Unsplash
Why is it important?
As energy systems become more complex and reliant on renewable sources, AI is seen as a promising tool to manage this transition. But the rush to adopt AI can overlook serious concerns—like fairness, cybersecurity, and whether AI decisions can be trusted or explained. This study is the first to bring together these technical and social concerns into one clear framework. By doing so, it helps researchers, policymakers, and energy professionals understand not just what AI can do, but what needs to be considered to use it responsibly and effectively in the power sector.
Perspectives
Working on this article was especially rewarding because it brought together my long-standing interests in sociotechnical systems, innovation policy, and the ethical dimensions of emerging technologies. As someone who studies the intersection of technology and society, I was particularly struck by how fragmented the conversation around AI in energy has become. I hope this paper helps spark more integrated thinking—not just about what AI can do, but about how we should be using it. If it encourages even a few readers to approach AI in power systems with more nuance, caution, and creativity, then it will have been well worth the effort.
Dr. Robert Edgell
SUNY Polytechnic Institute
Read the Original
This page is a summary of: AI in power systems: a systematic review of key matters of concern, Energy Informatics, June 2025, Springer Science + Business Media,
DOI: 10.1186/s42162-025-00529-1.
You can read the full text:
Contributors
The following have contributed to this page
